Monitoring passing trains, system and method

The present application is a U.S. National Stage application under 35 USC 371 of PCT Application Serial No. PCT/EP2021/074980, filed on 10 Sep. 2021, which claims priority to EP patent application No. 20195920.2, filed on 14 Sep. 2020, the entirety of each of which is incorporated herein by reference.

The invention lies in the field of monitoring railway networks and particularly in the field of monitoring rolling units on railway networks. The goal of the invention is to provide a method for monitoring a passing train on a railway network. More particularly, the present invention relates to a system for monitoring and detecting the speed of a passing train and the health status of components of the passing train, a method performed in such a system and corresponding use of a system.

Railroad, railway or rail transport has been developed for transferring goods and passengers on wheeled vehicles on rails, also known as tracks. In contrast to road transport, where vehicles run on a prepared flat surface, rail vehicles (rolling stock) are directionally guided by the tracks on which they run. Tracks commonly consist of steel rails, installed on ties or sleepers and ballast, on which the rolling stock, usually provided with metal wheels, moves. Other variations are also possible, such as slab track, where the rails are fastened to a concrete foundation resting on a subsurface.

Rolling stock in a rail transport system generally encounters lower frictional resistance than road vehicles, so passenger and freight cars (carriages and wagons) can be coupled into longer trains. Power is provided by locomotives, which either draw electric power from a railway electrification system or produce their own power, usually by diesel engines. Most tracks are accompanied by a signaling system. Railways are a safe land transport system when compared to other forms of transport. Additionally, railways are capable of high levels of passenger and cargo utilization and energy efficiency but are often less flexible and more capital-intensive than road transport, when lower traffic levels are considered.

Railway operations require careful monitoring and control of the conditions of the railway infrastructure to ensure passenger safety and reliable service. Many sensors are used to monitor and obtain data from different infrastructural component of the railway network, which may be used to ensure the integrity of the service and identify possible sources of malfunction. Such sensors allow for data collection and analysis and ensure safer operations of railways. Various sensors can be placed directly on trains, on tracks or nearby, at train stations and/or on platforms, and generally in the overall vicinity of the railway.

With the increase in rail traffic, rail system is under increasing pressure to keep the trains running on time and for longer. Safety, availability and reliability are the main components of a comfortable rail traffic. A system and method for analyzing the rail related data which fully integrates the type of the train, their location speed will help in understanding delays, infrastructural malfunctioning, etc. The International Union of Railways (IUR), the Community of European Railways (CER), the International Union of Public Transport (IUPT) and the Union of European Railway Industries (UNIFE) have all agreed, within the White Paper for European Transport, to attempt to increase the market share of goods traffic on rail from 8% in 2001 to 15% in 2020 (European Union, 2011). This will of course lead to an increase in railway traffic hence number of trains. Knowing the speed of a passing train will help in establishing the state of both the train itself and of the track, as well as knowing ancillary information about the train, such as its location, ETA, collision susceptibility, etc.

As the number of trains increases so will the data one can use from them. For example, the vibrations induced by the motion of the train via the interaction between wheel and rail tracks. This vibrational data can be used to extract a plurality of information, for example, rail and track bed condition, vehicle suspension, wheel condition, speed, weight of the vehicle, material used in tracks, depth to water table, frost depth, type of the vehicle, etc.

A system and/or a method which will be able to analyze this vibration signal will not only be able to provide the information about the wheels and the rail tracks but also the vehicle passing by and the traffic associated with the vehicle. As a person skilled in art may now infer that the nature of vibrations and the associated data will differ depending on the point where they are recorded. A few vibration/oscillation-based studies of railway related data have been done.

EP1274979B1 relates to a method for monitoring the travelling behavior of rail vehicles, according to which an oscillation behavior of at least one vehicle component is monitored by detecting at least one oscillation pattern and comparing the same with at least one reference oscillation pattern, whereby a natural oscillation of at least one vehicle component is monitored. The invention also relates to a device for monitoring the travelling behavior of rail vehicles, whereby at least one oscillation pick-up is mounted on at least one vehicle component. To this end, means are provided for evaluating the signal pattern, which is supplied by at least one oscillation pick-up, whereby characteristic values of the oscillation patterns of the at least one vehicle component are detected and compared with reference characteristic values of the oscillation patterns of a natural oscillation of the vehicle component.

CN102343922 provides an on-line monitoring system for vibration characteristics of a rapid railway turnout based on a wireless sensor network and relates to the technical field of safety monitoring of rapid railway infrastructures. The wireless sensor network serves as the core of a special system. The on-line monitoring system comprises a data monitoring unit for front-end three-shaft acceleration wireless sensor, a front-end data collecting unit for the wireless sensor network and a server terminal, wherein the data monitoring unit for front-end three-shaft acceleration wireless sensor is used for on-line acquiring a vibration data of a rapid turnout when a train passes by and sending the vibration data in a wireless mode; the front-end data collecting unit for the wireless sensor network is used for receiving the data sent by the data monitoring unit in real time and collecting and transferring the data; and the server terminal is used for receiving the data from the front-end data collecting unit, permanently storing the data, analyzing and calculating to obtain a train speed and a load condition according to the acceleration data, comparing the train speed and the load condition with a historic statistical data, and prompting and alarming for parameters which deviate from the historic statistical data and exceed a certain scope, thereby supporting the safety running of the rapid turnout. Combined with a conventional test and mechanical analysis method, the on-line monitoring system can be used for monitoring the rapid railway turnout and providing a data basis for maintaining and design optimizing of the turnout.

In light of the above, it is an object of the present invention to overcome or at least alleviate the shortcomings of the prior art. More particularly, it is an object of the present invention to provide a method and system for analyzing vibration signal data to related to properties of a passing train on a railway network.

These objects are met by the present invention.

In a first aspect, the invention relates to a method for analyzing properties related to a passing train on a railway network, the method comprising collecting a first sensor data of railway at a first time via a least one sensor arranged on the railway network, processing the first sensor data via at least one processing component to generate at least one processed sensor dataset, and estimating a value of at least one property related to the passing train based on the at least one processed sensor data to generate at least one property estimation of the passing train.

The value of the at least one property of the passing train may comprise a speed of the passing train, and wherein the at least one property estimation may comprise at least one speed estimation of the passing train.

In one embodiment, the method may further comprise predicting the speed of the passing train based on the at least one sensor data to generate at least one speed prediction of the passing train.

Furthermore, the method may comprise may further comprise connecting the at least one sensor to at least one server.

In one embodiment, the method may comprise retrieving at least one vibration signal.

The method may further comprise extracting at least one Power Spectral Density (PSD) data set from the least one vibration signal.

In one embodiment, the at least one vibration signal returns a PSD as at least one-dimensional vector of a fixed size.

The at least one acceleration signal may comprise a signal frequency signal and a sampling rate, wherein the signal frequency may be between 0 and 10,000 Hz, preferably between 50 and 8,000 Hz, more preferably between 100 and 5,000 Hz, and wherein the sampling rate may be between 0 and 20 kHz, preferably between 0.1 and 8 kHz, more preferably between 1 and 5 kHz.

In one embodiment, the method may comprise extracting at least one Mel spectrogram from the least one vibration signal.

The method may comprise mapping the data to generate a mapped data set.

Furthermore, the method may comprise curating the at least one estimation of the train speed based on the mapped data.

In one embodiment, the method may comprise bidirectionally connecting the at least one sensor to the at least one server.

Moreover, the method may comprise connecting the at least one server with the at least one processing.

In one embodiment, at least one of the at least one server may comprise at least partially one of the least one processing component.

The method may comprise facilitating the at least one sensor with a sensor processing component.

In one embodiment, the at least one processing component may comprise a memory component configured to store at least one of the first sensor data, and the at least processed sensor data.

The sensor processing component may comprise a sensor memory component configured to store at least one of the first sensor data, and the processed sensor data.

In one embodiment, the method may further comprise retrieving from the at least one sensor at least one of the first sensor data, and at least one sensor ID, and supplying to the at least one server at least one of the first sensor data, and the at least one sensor ID, wherein the at least one sensor ID may be related to the at least one sensor.

The at least one vibration signal may comprise at least one of: at least frequency data, at least displacement data, at least velocity data, at least acceleration data.

The method may further comprise automatically generating at least one acceleration trace associated with the first sensor data.

In one embodiment, method may comprise automatically transmitting to the at least one processing component at least one of the first sensor data, and the at least processed sensor data.

Moreover, the method may comprise pre-processing the via the at least processing component the first sensor data.

In one embodiment, step of pre-processing may further comprise at least one of: flagging at least one noisy component of the first sensor data, removing at least one exponential wakeup, cutting off the edge of the at least one acceleration trace, stretching the at least one first sensor data to a pre-determined size, representing the at least one first sensor data as a time-frequency spectrogram.

The step of flagging may comprise automatically deleting at least a pre-determined section of the acceleration trace.

In one embodiment, the method may further comprise the step of automatically calculating the pre-determined section according to a type of the passing train.

Moreover, the step of flagging may further comprise altering the at least one acceleration trace when a root mean square (RMS) value of the acceleration may be lower than a threshold value.

Additionally or alternatively, the step of removing the exponential wakeup may comprise alteration of an automatically pre-calculated number of acceleration trace/s from the acceleration trace.

In one embodiment, the method may further comprise the step of automatically identifying the pre-generated number preferably by fitting exponential curve to the acceleration trace and differentiating between a real signal and a wakeup curve.

The method may further comprise the step of cleaning at least one additive noise, preferably by implementing a wiener filter.

The method may further comprise dynamically extracting at least one temporal and spectral content to standardize an output to a fixed size, and automatically converting the at least one acceleration trace to at least one time-frequency spectrogram.

The method comprising scaling the at least spectrogram value within a pre-determined region.

In one embodiment, the method comprising generating at least one spectrogram parameter preferably using hyperparameter optimization on at least one pre-determined truth dataset.

The method may further comprise facilitating the at least one processing component with at least one neural network component.

In one embodiment, the method may comprise feeding into the at least one neural network component at least one pre-processed dataset of the first sensor data.

The at least one neural network may comprise at least one convolutional neural network layer.

In another embodiment, the method may comprise automatically extracting at least one feature map from at least one of the first sensor data, and the at least processed sensor data.

The method may comprise using the outcome of the pre-processing step for predicting the speed of the passing train.

In one embodiment, the method may further comprise using the outcome of the pre-processing for embedding the at least one acceleration trace in the feature map.